Jan., 1913 CONCEALING AND REVEALING COLORATION OF ANIMALS 9 likely to survive, and the others to perish. Another species may be varying in the direction Of revealing coloration, but at the same time developing some other element of safety which far outweighs the disadvantage of revealing coloration, and thus would survive. Most naturalists who have studied the two in the field will probably agree that the crow, whose color under most circumstances cannot be considered at all concealing , is, because of its intelligence, alertness and ac- tivity, quite as capable of escaping its natural enemies, armed only with natural weapons, as is the ptarmigan, which affords one of the best examples of conceal- ing coloration. The law of compensation runs all through nature, animate as well as inani- mate, and cannot be ignored by naturalists. In the natural course of things, a more or less revealingly colored animal would be expected' to develop its alert- ness, speed or some other factor of safety, to a greater extent than its better-con- cealed neighbor. In fact, this seems to be the actual result in certain familiar cases, though not at all in other instances, so far as we may judge. There is ?ome reason for the supposition that reliance upon concealment in many cases enables an enemy to approach very closely before escape is attempted. Thus conceal- ing coloration, reacting upon mental and physical activity, may possibly some- times be an actual disadvantage, if its concealing effect be known to its possessor, of which we cannot very well be certain until we can persuade the animals to tell U s . It is possible, if not probable, that many ofher causes besides the need of con- cealment contribute to the coloration of animals. Mr. Beebe's experiments upon the direct effect of moisture, dryness, heat, cold, diet, etc., upon animal colors are well known and enlightening, whatever the final conclusions may be. Some other facts bearing upon this phase of the problem, generally known to professional zoologists, are often ignored, and may not be so well known to amateurs. For example: (a) The general possession of dark skins by tropical animals, which is not confined to tropical human beings and which is possibly not due merely to the direct blackening of the skin by the sun ("tanning"), but to pigmentation for protection of the living tissues from the destructive action of intense light, ex- cessive heat, or both. (b) The difference between different colors in their power to radiate heat, which may require animals of different habitat to be differently colored without reference to concealment. Thus such colors would be protective, yet not concealing. "Protection" is a broader ?term than "concealment," and the two should not be used synonymously. (c) The difficulties found in attempting to introduce white-skinned animals (swine) into certain regions, which may be due to the deleterious effect of intense light. (d) The difference in the color of domestic horses under different climatic conditions, possibly due to temperature ?r light, or moisture, a matter now under investigation. Professor Chas. F. Woodruff has recently discussed this subject (Science, n. s., xxxv, April ?2, ?9?2, pp. 59?-593). The recent change of color of the linnet introduced into Hawaii may be due to some such cause. If no animal were in need of concealment, it is probable that species would still differ in color and color pattern. Among the smaller animals many species differ minutely in color and color pattern, yet so slightly that the differences are scarcely discernible, exc.ept by direct comparison, with specimens in hand. These differences can hardly be for concealment. Sometimes colors are the result ?f refraction of light due to physical structure, as in pearls, opals and iridescent feathers of birds. Sometimes they are the result of chemical composition, as in rubies, emeralds, sandstones, limestones, etc. In the cases just mentioned, except
